EJIL (2017), Vol. 28 No. 1, 233–240 doi:10.1093/ejil/chx005
© The Author, 2017. Published by Oxford University Press on behalf of EJIL Ltd. All rights reserved. For Permissions, please email: journals.permissions@oup.com
Law Beyond the State: A Reply
to Liam Murphy
Samantha Besson*
Voluntarism about the very sources of international law is incoherent, but there is a better way to interpret this view. The sources of international law being what they are, legal obligations are never imposed on states without their consent. ... The claim ... is false (though perhaps it once was true), but it is perfectly coherent, and it is not absurd to wish that it were (still) true.
Liam Murphy, ‘Law Beyond the State: Some Philosophical Questions’1
1 State Consent, Legal Positivism and the Duty to Obey
International Law
Liam Murphy was among the first analytical legal philosophers to venture into discus-sions of the philosophy of international law.2 His contribution to the field is also one of the most illuminating.3 His new piece ‘Law beyond the State: Some Philosophical Questions’ is a welcome consolidation and an update of his views on the subject. One of Murphy’s most interesting arguments in this piece, besides his superb reconstruc-tion of H.L.A. Hart’s views about the existence of an internareconstruc-tional legal system and the best way to understand them in relation to Hans Kelsen’s (section 2), is his discus-sion of the role of state consent in international law-making. It is also the one I would like to take issue with in this short reply by linking it to his considerations on legal positivism (section 1) and the ‘duty to obey global law’ (section 5).
Murphy is right to argue in the first section of his article that legal positivism is best kept separate from voluntarism in international law. The latter’s identification with legal positivism is largely accidental and, at best, accounted for on historical and politi-cal grounds.4 Even if it is descriptively correct to consider that state consent plays an
* Professor of Public International Law and European Law, University of Fribourg, Switzerland. Many thanks to Jose Luis Martí for our many disagreements and to Matthieu Loup for his editorial assistance. Email: samantha.besson@unifr.ch.
1 Murphy, ‘Law Beyond the State: Some Philosophical Questions’, in this issue, 203, at 205.
2 See Besson, ‘Moral Philosophy and International Law’, in F. Hoffmann and A. Orford (eds), Oxford
Handbook of International Legal Theory (2016) 385.
3 See, e.g., Murphy, ‘International Responsibility’, in S. Besson and J. Tasioulas (eds), The Philosophy of
International Law (2010) 299; L. Murphy, What Makes Law (2014), ch. 8.
important role in the positing of international law, this does not make the grounds of international law a matter of will. Considering consent as a criterion for the validity of international law would be question begging; no state has consented to the prin-ciple of law according to which treaties (that states have consented to) are a source of law – that is, pacta sunt servanda.5 In short, as Murphy puts it, ‘voluntarism about the very sources of international law is incoherent’, but this should not be a concern for international legal positivism because voluntarism is not ‘implied by the idea that the grounds of law are matters of fact’.6
While I agree with Murphy that state consent cannot account for the validity of international law, I would like to argue, as I have elsewhere, that state consent plays an important normative role in international law-making.7 After all, international legal obligations are never imposed on states without their consent, and this requires a justifi-cation. Of course, unlike many authors who have argued against the conceptual connec-tion between state consent and validity in internaconnec-tional law, and then consider consent as an unhappy resilient feature of the international law practice that one should be pragmatic and resigned about, Murphy does take seriously the role of states’ consent in accounting for their obligations under international law. According to him, ‘the fact that most states have in fact consented to most of international law is ... very important’.8 He even concedes that linking international legal obligations to state consent is a ‘perfectly coherent’ claim and perhaps one that was once ‘true’.9 Nevertheless, Murphy adds that today the claim that ‘legal obligations are never imposed on states without their consent’ is ‘false’.10 He discusses this claim in the fifth section of the article. However, the author addresses what he calls ‘the argument from consent’ only as a ground for the duty to obey international law. This is regrettable because there are many other ways to argue for the normative role of state consent than to make the long discredited argument that consent provides a justification for the authority of international law.11
In short, while I agree with Murphy that consent is neither a criterion for the validity of international law nor a ground for its legitimacy, one needs to consider alternative justifications for the role of state consent with respect to the legitimacy of international law. Murphy’s argument against consent as a ground for the legiti-mate authority of international law is both descriptive or ‘factual’ and normative.12 As I will explain, however, the proposed descriptive rebuttal fails to convince, and the normative critique is incomplete. It should be complemented, I will propose, by a democratic argument for state consent as an exception to the legitimate authority of international law.
5 Ibid. 6 Ibid.
7 See Besson, ‘State Consent and Disagreement in International Law-Making: Dissolving the Paradox’, 29
Leiden Journal of International Law (2016) 289, at 298–299, 303–304.
8 Murphy, supra note 1, at 230. 9 Ibid., at 205.
10 Ibid.
11 See Besson, supra note 7, at 300–302. 12 Murphy, supra note 1, at 230.
2 Debunking the Descriptive Case against State Consent
Murphy is very cautious about the descriptive part of his argument against state con-sent in international law-making. He concedes that the ‘factual basis of the concon-sent argument has never been its gravest weakness’.13 He adds that the case for it is ‘not laughable’ and, later on, that ‘most states have in fact consented to most of interna-tional law’.14 All the same, he claims, ‘the case for the actual fact of consent’ has not been made out. Of course, Murphy is not alone in making this claim. It has become common for international lawyers to observe or predict the erosion of state consent in international law-making.15 Importantly, however, the changes usually identified are mostly located at the periphery of international law16 – indeed, the examples most mentioned are the rise of soft law and the prevailing role of unilateral decisions by powerful states.17 Although there are clearly new non-consensual means of interna-tional cooperation, none of them is (yet) regarded as a source of internainterna-tional law stricto sensu. More importantly, the latter remains mostly consent based. This is the case with the most important sources of international law – that is, treaties and cus-tomary international law. Thus, if there is a threat to the role of state consent in inter-national law-making it is external, rather than internal, to interinter-national law.
Murphy puts forward three elements to ‘undermine the claim that international law imposes no obligation without consent’: the existence of jus cogens norms that states should not derogate from even if they disagree; the fact that custom binds every new state qua state and independently of its consent; and, finally, the legislative role of international organizations whose law binds states even against their consent.18 Additionally, the author indicates that he does not find the ‘doctrine’ of persistent objection ‘terribly secure’ and, accordingly, a convincing argument for the role of state consent in customary interna-tional law-making, although he does not say why and seems to endorse it all the same.19
Starting with the last argument and going up the list, it is not surprising that custom-ary international law may seem, prima facie, to bind some states without, or even against, their consent; its formation depends on a general practice only, but its authority extends to all states. Importantly, however, this authority only extends to a state provided it has not objected, expressly and persistently, to the emerging consensus. Each state may there-fore dissent and, in so doing, withdraw the consent it was otherwise giving tacitly.20 As a result, customary international law-making combines tacit consent in the converging practice of states and explicit dissent in their possibility to object to that practice through a persistent objection. To this extent, like international treaties, customary international
13 Ibid. 14 Ibid.
15 See, e.g., Tomuschat, ‘Obligations Arising for States without or against Their Will’, 241 Collected Courses
of the Hague Academy of International Law (1993) 195.
16 See, e.g., Krisch, ‘The Decay of Consent: International Law in the Age of Global Public Goods’, 108
American Journal of International Law (AJIL) (2014) 1, at 2, 26ff, 34.
17 See, e.g., J. Pauwelyn, R. Wessel and J. Wouters (eds), Informal International Law-Making (2012). 18 Murphy, supra note 1, at 230.
19 Ibid.
law cannot impose obligations on states without their consent.21 This rejoinder should also placate Murphy’s second point about every new state being bound by customary international law. This claim is false because each state may oppose its persistent objec-tion. The self-determination of newly created states was actually one of the justifications for the introduction of the possibility to waive a persistent objection in the 1960s.
In reply to Murphy’s third critique – that is, the legislative role of international organizations – one should stress that the authority of their internal law still relies indirectly on their constitutive treaties and, hence, on state consent or, at least, on states’ subsequent converging practice.22 The same may be said about the develop-ment of international adjudication. This leaves us with Murphy’s first objection – that is, the emergence of jus cogens norms. What characterizes jus cogens norms, however, is their normative stringency but not their sources. Of course, the latter have to be such that their stringency can be absolute. However, practice shows that jus cogens norms arise, when they do, from treaties as much as from customary international law. This is also what Article 53 of the Vienna Convention on the Law of Treaties indicates when it states that a jus cogens norm is a norm ‘accepted and recognized by the international community of states’ as such.23 As a result, a jus cogens norm cannot arise without the consent of the states it binds.
As I will explain in the next section, however, the justification of the opposability of state consent to the legitimate authority of international law is democratic, and this, in turn, gives it inherent democratic limits. In practice, these limits usually take the shape of non-discrimination rights, on the one hand, and of absolute or minimal human rights duties or other jus cogens norms, on the other. This accounts for why valid reser-vations, for instance, cannot be made to absolute duties in human rights treaties,24 for how international succession law excludes applying the tabula rasa principle to human rights treaties, or for how third-party obligations arise from human rights treaties. In order to fully grasp the inherent limits to state consent, it is important to start by mak-ing the normative case for its role in the justification of obligations under international law. It is that very normative prong of the argument for state consent that is regarded as the weakest by Murphy, and the time has come to turn to a discussion of his critique.
3 Nuancing the Normative Case against State Consent
Prima facie at least, Murphy’s normative argument against state consent is very strong. He rightly explains why consent cannot be a ground for the duty to obey inter-national law. To do so, Murphy relies on the classical argument developed in political theory against consent as a ground of political and legal authority, although he does
21 See also International Law Commission, Third Report on Identification of Customary Law, by Michael
Wood, Special Rapporteur, UN Doc. A/CN.4/682, 27 March 2015, at 61–69.
22 See also Klabbers, ‘Law-Making and Constitutionalism’, in J. Klabbers, A. Peters and G. Ulfstein (eds), The
Constitutionalization of International Law (2009) 81, at 100, 114.
23 Vienna Convention on the Law of Treaties 1969, 1155 UNTS 331.
24 See, e.g., Klabbers, ‘The Validity and Invalidity of Treaties’, in D.B. Hollis (ed.), The Oxford Guide to Treaties
not make this argument himself besides a mention in passing.25 As Joseph Raz pointed out early on, while consent may account for why promises are morally binding, it does not justify why the law binds and, more specifically, give us moral reasons for action. Indeed, one may consent to do wrong, while one cannot have a moral reason to do so.26 As a result, consent cannot be a primary ground for the law’s authority unless the law respects autonomy and satisfies an independent test of legitimacy. Importantly, it is not only the plausibility of actual or express consent to the law that is at stake in this rebuttal, since, as Murphy emphasizes, it would actually be plausible in international law in contrast to domestic law, but also the conceptual-normative inability, even qua hypothetical or tacit consent, to justify the law’s authority.27
Murphy mentions the distinction between treaties and soft law agreements to illustrate this point.28 While the former bind as law, the latter (which are not (yet) law) still bind as promises, thereby confirming that the duty to obey the law cannot be consent based. This is an interesting example to the extent that, as I explained before, soft law is often mentioned as evidence of the lack of relevance of state consent in international relations, whereas its development should lead to the exact reverse conclusion – soft law is pre-dominantly based on (gentlemen’s) agreements between states and actually binds states because they have consented to it. Of course, as Raz recognizes, consent may constitute an additional reason to respect the law or have trust in it. Consent may contribute to enhance the de facto authority of international law by strengthening respect for it in prac-tice.29 This is also something Murphy emphasizes when he refers to Allen Buchanan’s argument regarding the subjective legitimacy of international law and accountability.30
Once consent is severed from the potential grounds of legitimacy of international law, the next question, of course, is what those grounds actually are. This question is not fully settled by Murphy and needs not be settled fully here either.31 In short, I have argued elsewhere for a revised Razian conception of legitimate authority whereby international law binds to the extent that the (content-independent and exclusion-ary) reasons it provides enable its subjects (states and individuals in those states) to comply better with the reasons that apply independently to them (service concep-tion). The grounds for the authority of international law may range from volitional, expressive and epistemic to coordinative reasons. In the revised account I have pro-posed, however, the main basis for the demand for legitimate authority is reasonable disagreement about how to structure common action when such action is required morally. Accordingly, the justification of legitimate authority lies mostly in the coor-dinating ability of international law in circumstances of reasonable disagreement.32
25 Murphy, supra note 1, at 230, n. 102. In international law, see Buchanan, ‘The Legitimacy of International
Law’, in S. Besson and J. Tasioulas (eds), The Philosophy of International Law (2010) 79.
26 See J. Raz, Ethics in the Public Domain (1995), at 80–94, 355–369. 27 Murphy, supra note 1, at 229–230.
28 Ibid., at 230.
29 See Raz, supra note 26, at 368–369. See also Besson, ‘The Authority of International Law: Lifting the
State Veil’, 31 Sydney Law Review (2009) 343, at 352, 371–372.
30 Murphy, supra note 1, at 230–231. A. Buchanan, Justice, Legitimacy, and Self-Determination: Moral
Foundations for International Law (2007).
31 Murphy, supra note 1, at 231. 32 See Besson, supra note 29, at 351ff.
Moreover, and for the same disagreement-related reasons, this account accommo-dates the importance of public and egalitarian, and especially democratic, authority by recognizing coordination as a ground for general, and not just piecemeal, authority.
Against this background, it becomes clear how state consent can contribute to rein-forcing the salience of the coordinating option – for instance, in international treaty making.33 It amounts to a public method for the creation of content-independent rea-sons to obey international law in the absence of a centralized lawmaker.34 This is par-ticularly important in the circumstances of substantive and epistemic disagreement that prevail in international law.35
The coordination dimension and generality of the duty to obey international law are also emphasized by Murphy.36 Unlike him, however, I think it is a feature that accounts of the authority of international law can share with domestic ones. Indeed, it is not clear why the weakness of the international enforcement mechanism should ground an instrumental duty of obedience for states that is more general than individual ones domestically.37 International law is primarily interpreted and enforced domestically, and, to this extent, its enforcement mechanisms are not really weaker than domestic ones. Moreover, individuals are the duty bearers of international law as much as of domestic law, and these two sets of legal duties are not easy to disentangle. Finally, the fact that the domestic duty to obey is political (or, at least, more political than the international one) does not make any difference in terms of generality, at least from a democratic perspective.38 On the contrary, its publicity and egalitarian dimensions require its generality.
4 Complementing the Normative Case for State Consent
While I agree with Murphy that state consent does not provide a ground for the duty to obey international law, I think its role goes beyond enhancing general respect and compliance with international law. The central role of state consent in contemporary international law-making is justified democratically, as I have argued elsewhere, and even more so in the circumstances of international reasonable disagreement.39 At first, an argument for ‘democratic state consent’ may sound paradoxical. Indeed, domes-tically, the role of consent has long been disparaged from a democratic perspective precisely because of reasonable disagreement. Actually, the equality-based justification of democracy accounts for majority voting instead of unanimity, thus making consent even less relevant procedurally in a democracy.
33 See also ibid., at 353, 371–372.
34 See Christiano, ‘Climate Change and State Consent’, in J. Moss (ed.), Climate Change and Justice (2015) 17. 35 See Pellet, ‘The Normative Dilemma: Will and Consent in International Law-Making’, 12 Australian
Yearbook of International Law (1988–1989) 22, at 45.
36 Murphy, supra note 1, at 231. 37 Ibid.
38 Ibid.
Importantly, however, democratic considerations in international law should not be conflated with domestic ones.40 First of all, states should not be treated as equal indi-vidual members of an international democratic polity in the same way as indiindi-viduals are equal members of a domestic democracy. There is no democracy of states, but there should not be one either. States cannot, and should not, be too readily identified with individual subjects whose basic equality actually justifies democracy; the same justification could not arise from the equality of states. Second, the conditions of a global democracy for individuals in which states would merely act as their officials are not given either. Indeed, in the absence of the egalitarian pre-conditions for global democracy, and, in particular, of equal and interdependent stakes shared by all of us internationally, there cannot (yet) be a global democracy.41 As a result, the legitimacy of international law cannot, and should not, strictly speaking, be democratic.
All the same, the importance of democratic legitimacy domestically implies that we should try to find a way to respect domestic democracy in the way we make interna-tional law and, especially, in the political equality of the members of democratic states. After all, to quote Thomas Christiano, states remain ‘the most important institutional mechanism for making large scale political entities directly accountable to people’ and, as a result, the sole forum of democracy.42 It should be clear, therefore, that the way to link international law-making processes to domestic democratic legitimacy is to respect the equality of each democratic state qua statespeople.43
The way to protect the equality of states (people) qua collective equality is to con-sider their consent as a requirement in international law-making. The requirement of the equal consent of states enables small and weak states to resist the domination and the hegemony of large and powerful states or the coalition of states.44 In turn, states’ equal consent protects the individual members of those states’ right to an equal voice in the collective decision-making process they are participating in through their states.45 Importantly, considering democratic state consent as a requirement of inter-national law-making does not mean considering it as a ground for its legitimacy. As we saw before, it simply cannot be such a ground, whether tout court or in a democratic context. As a matter of fact, the legitimacy of international law is not democratic, strictly speaking.46 Its legitimacy is justified on the grounds of coordination in circum-stances of reasonable disagreement. All the same, it is precisely because reasonable disagreement among states is widespread and persistent in international law, on the one hand, and because of the centrality of democratic states in making international power accountable to their people, on the other hand, that democratic state consent should work as an exception to the prima facie legitimate authority of international
40 See Besson, supra note 29, at 368–370.
41 See also Christiano, ‘Democratic Legitimacy and International Institutions’, in S. Besson and J. Tasioulas
(eds), The Philosophy of International Law (2010) 119.
42 Christiano, supra note 34. 43 See Christiano, supra note 41.
44 See, e.g., Klabbers, supra note 22, at 114. 45 See Christiano, supra note 34.
law. In these conditions, the way in which democratic state consent ties in is merely as an exception to the legitimate authority of international law that is justified on other grounds.
While democratic state consent is an important dimension of the legitimacy of international law, there are inherent limits to it. The first set of limits pertains to its democratic justification and the protection of the democratic statespeople. Those lim-its amount to the protection of the basic conditions of democracy – that is, minimal political equality and absolute human rights duties. A second set of limits inherent to democratic state consent have to do with consent itself. There are at least three of them: the free and unconstrained nature of state consent; its fairness and egalitarian features; and its informed and unbiased nature.47 Even within those constraints, there are at least four critiques one may make to democratic state consent from a demo-cratic theory perspective.48 First of all, state consent does not pay enough attention to individual equality and to states’ proportional demography. Second, it is largely veto-centred and does not encourage sufficient deliberation among democratic states. Third, the international actions of states’ representatives are not necessarily submit-ted to internal democratic control. Finally, state consent only protects democratic self-determination when the state is democratic.
The first three critiques may be addressed through various internal and interna-tional reforms. This should be done, first of all, through enhancing proporinterna-tional rep-resentation and decision making in international fora such as international courts and legislative assemblies; second, through developing transnational comparison and deliberation in international legal interpretation and determination; and, third, through submitting the internal approval not only of international treaties but also of international negotiation mandates of the government to parliamentary procedures that equate internal legislative ones. The fourth critique is the hardest to address, how-ever. Indeed, either the state consent exception only benefits democratic states, at the expense of equality with individuals in non-democratic states, or, if it is generalized to all states, it enables the agenda of non-democratic states to dominate democratic ones, again at the expense of individual equality albeit within democratic states. The least objectionable answer therefore is to restrict the right to invoke or oppose one’s state consent to democratic states only.49 This is the price of the minimal legitimacy stan-dards that arise from state practice in international law and bind states legitimately even without their consent.50
47 See Christiano, supra note 34.
48 See Martí, ‘Sources and the Legitimacy of International Law’, in S. Besson and J. d’Aspremont (eds),
Oxford Handbook on the Sources of International Law (forthcoming 2017).
49 See also Christiano, supra note 34.
50 See also Buchanan, ‘Reciprocal Legitimation: Reframing the Problem of International Legitimacy’, 10
